Information technology systems (e.g., servers, storage systems, switches, etc.) generate large amounts of heat during operation. This heat must be removed from these systems in order to maintain safe operating temperatures. The components within an overheated system generally exhibit a shorter maximum lifespan and may give sporadic problems. Cooling of these systems is typically accomplished by way of forced convection cooling. In these systems, cooling fans are used to force air to flow over the heat generating devices.
An individual cooling device may not be sufficient to supply the cooling needs of a whole system. In such a situation, multiple cooling devices may operate simultaneously to provide sufficient cooling of the system. In some cases, more fans than are necessary may be used, a technique known as redundant cooling. When multiple cooling fans are used, even in cases where there is no cooling redundancy, a single cooling fan may fail, but the others may continue operating in order to prevent overheating. The system may increase the output of the remaining cooling fans to compensate for the failed cooling fan and to notify an operator that a cooling device unit has failed.